将已经四舍五入的浮点数字符串化，然后将其转换回数字会产生进一步的精度误差吗？

Question

Accrodading to Is floating point math broken? , I know 0.1+0.2 results in 0.30000000000000004 instead of 0.3 because of floating point errors. However, what if I copy "0.30000000000000004" from alert(0.1+0.2) and write Number("0.30000000000000004"), would it still equal to 0.1+0.2?

Or in general:

let a=float_number_1 + float_number2;
let b=Number((float_number_1 + float_number2).toString());

Does a always exactly equal to b?

Answer 1

Java's default conversion of a number to a string is specified to produce enough digits to uniquely distinguish the number from adjacent representable values. In other words, it is deliberately designed to produce a string that can be converted back to a number to produce the original value.

So, if you convert from a float to a string to a float or from a double to a string to a double , you will get the original number. If you convert a double to a string to a float , you may get a different number.